dataeval.shift.DriftMVDC¶

class dataeval.shift.DriftMVDC(n_folds=None, chunk_size=None, chunk_count=None, threshold=None, config=None)¶

Multivariant Domain Classifier

Parameters:¶

n_folds : int, default 5¶

Number of cross-validation (CV) folds.

chunk_size : int or None, default None¶

Number of samples in a chunk used in CV, will get one metric & prediction per chunk.

chunk_count : int or None, default None¶

Number of total chunks used in CV, will get one metric & prediction per chunk.

threshold : Tuple[float, float], default (0.45, 0.65)¶

(lower, upper) metric bounds on roc_auc for identifying drift.

config : DriftMVDC.Config or None, default None¶

Optional configuration object with default parameters. Parameters specified directly in __init__ will override config defaults.

Examples

Using configuration:

>>> config = DriftMVDC.Config(n_folds=10, threshold=(0.4, 0.6))
>>> detector = DriftMVDC(config=config)

fit(x_ref)¶

Fit the domain classifier on the training dataframe

Parameters:¶

x_ref : ArrayLike¶

Reference data with dim[n_samples, n_features].

Return type:¶

DriftMVDC

predict(x)¶

Perform inference on the test dataframe

Parameters:¶

x : ArrayLike¶

Test (analysis) data with dim[n_samples, n_features].

Return type:¶

DomainClassifierDriftResult

Classes¶